Quote:
Originally Posted by Bobsterz
I normally go from Word to HTML (that I clean up, mostly with mass search and replace, but it can become arduous in places) and then to Kindle.
I like the process, because I have exact control, Word faithfully reproduces the external links and TOC, and my use of the outline view and styles result in excellent formatting once I put in my own CSS that uses the styles already created.
|
Great. Sounds like you're already 95%+ there.
Styles, and learning how to use them properly, is the #1 important step for designing clean documents.
I wrote about Styles + Word (or LibreOffice)->EPUB in detail last year in:
"eBook Formatting in Sigil" (Posts #46, #50, but especially #52+#60 where I discuss the DOCX->EPUB tools).
Quote:
Originally Posted by Bobsterz
But what is the current thinking about going straight from Word to Kindle? Is there a lot of back and forth, getting the look right? I'm getting the impression that things have changed and maybe it's a lot easier nowadays to bypass the HTML.
|
No. I would still go:
- DOCX->EPUB (or DOCX->HTML->EPUB)
- then EPUB->Kindle
EPUB gives you
full control over the final ebook file.
Feeding DOCX into Kindle Previewer, then generating a Kindle file out of that... who knows what crap/cruft would get carried along + not get converted properly:
- auto-numbered lists
- lists-that-don't-start-at-1
- footnotes/endnotes
- poems/lyrics
- blockquotes
- metadata
- [...]
The more complicated your book, the more chance of things going wrong.
And you'll always have to dig in there and do HTML/EPUB tweaks, because physical book ≠ ebook.
Quote:
Originally Posted by Bobsterz
P.S. Word's process for creating an index does not work for going into HTML (proper links are not produced), so this made for a lot of extra work, producing the index in HTML after the conversion. It wasn't awful, but it certainly takes longer. That's one more motive for asking about bypassing the HTML.
|
Indexes... and trying to link Indexes...
No.
As you found out, Word
does not export the proper index code to HTML.
The only tool I've found, which carries over Word Indexes to HTML, is Calibre. (Although I haven't messed with that functionality in a while.) But that has the enormous disadvantage of renumbering every entry 1->n in the order they appear.
So if you had:
Code:
cats, 100, 101
dogs, 99, 100
elephants, 50, 101
you'd get a whole mismmash:
Code:
cats, [1], [2]
dogs, [3], [1]
elephants, [4], [2]
What you'd have to do is manually recreate "Real Page Numbers" (RPNs), output the Index as plaintext, then manually regex the Index back together. A lot of that was written about in:
2020: "Create index on epub from printed book"
2020: "How to export indexes from indesign to epub"
And search for this in your favorite search engine:
Code:
"Real Page Numbers" tex2002ans site:mobileread.com
(Over the years, I've also written a ton about Indexes + Indexes in Ebooks... but I tend to stay ten thousand feet away from linking them. The usefulness in ebooks is debatable, because "a page" in a physical book ≠ "a screen" in an ebook [the linked info may be 1–4+ screens away].)
Quote:
Originally Posted by Bobsterz
P.P.S I'd love to hear from folks who take on jobs that involve HTML clean-up and prepping Word files for paperback conversion as well. I need to hire out my upcoming jobs.
|
Sure. Like Quoth said, there's:
... and then me (I've done 600+ ebooks, mostly Non-Fiction).
Not too sure of any other MobileRead users who do it professionally at that skill level.